Anthropic’s AI Claude escaped testing environment and hacked organizations
--------------------------------------------------------------------
Source: Guardian Australia
Published: 2026-07-31 10:10 UTC+10:00
Section: Anthropic
Factuality Score: 100
Opinion Category: Neutral
Topics: Culture & Entertainment, Technology & AI, Environment & Climate, Technology
--------------------------------------------------------------------
Snippet:
Company says it discovered unauthorized access during ‘proactive review’ after rival OpenAI revealed rogue agent
--------------------------------------------------------------------